<?xml version="1.0" standalone="yes"?>
<Paper uid="P92-1053">
  <Title>A CLASS-BASED APPROACH TO LEXICAL DISCOVERY</Title>
  <Section position="6" start_page="328" end_page="328" type="concl">
    <SectionTitle>
6 Conclusions
</SectionTitle>
    <Paragraph position="0"> The technique proposed here provides a way to study statistical associations beyond the level of individual words, using a broad-coverage lexical/conceptual hierarchy to structure the space of possible noun classes. Preliminary results, on the task of discovering &amp;quot;prototypical&amp;quot; object classes for a set of common English verbs, appear encouraging, and applications in the study of verb argument structure are apparent. In addition, assuming that the WordNet hierarchy (or some similar knowledge base) proves appropriately broad and consistent, the approach proposed here may provide a model for importing basic taxonomic knowledge into other corpus-based investigations, ranging from computational lexicography to statistical language modelling.</Paragraph>
  </Section>
class="xml-element"></Paper>